Skip to main content

Docs-as-code migrations

Confluence to Docusaurus migration

This is for SaaS teams that write product or developer documentation in Confluence and have outgrown it: the page tree grew by accretion, macros hide content from search engines and readers, and customer-facing docs share a tool built for internal notes. We convert the spaces you choose into a Docusaurus site in your Git repository, rebuild what the macros did, and redirect the old links so nothing your customers or agents bookmarked goes dead.

Price

From $4,500 (up to 150 pages), $9,500 (up to 500 pages), custom above

Response

Reply within 1 business day

After delivery

30-day fix window

Who it's for#who-its-for

  • Docs leads whose public documentation lives in a Confluence space next to internal pages, with permissions doing the job that structure should be doing.
  • Engineering leads who want API and developer docs beside the code, reviewed in pull requests and built by CI like everything else they ship.
  • Heads of support whose agents paste Confluence links into tickets every day and need those links to keep resolving after the move.
  • Operations and IT leads consolidating tools, who want customer-facing documentation out of the internal wiki and into a site the company fully controls.

Internal content usually stays where it is. The migration covers what readers outside the team need, and the inventory makes that boundary explicit before any conversion starts, so nobody discovers later that a runbook went public.

What you get#what-you-get

A Docusaurus site that reads as if it had been written for Docusaurus. That means dealing with the parts of Confluence that do not convert on their own:

  • Page content taken from the HTML space export or the REST API, which returns pages in Confluence's XHTML-based storage format, and converted into clean MDX with frontmatter.
  • Macros mapped to equivalents: info, note and warning panels to admonitions, expand macros to collapsible sections, code blocks with their language set, and includes and excerpts to reusable MDX partials.
  • Jira issue macros and other dynamic content replaced with static text or links, each one flagged in the inventory for your review.
  • Attachments and images moved into the repository, deduplicated, renamed sensibly and given alt text.
  • Internal links rewritten to the new paths, and every old Confluence URL mapped in a redirect list.
  • A restructured sidebar: the page tree rebuilt around reader tasks instead of the order in which pages happened to be created.
  • CI that fails the build on broken links and anchors, so the structure stays intact after handoff.

Confluence pages also carry markup that looks fine in the wiki and breaks everywhere else: nested tables, colored status text, images sized inline, headings used for emphasis. The conversion notes record how each pattern was handled, so your writers know what to expect in the new files.

How it runs#how-it-runs

  1. Scoping. You export the space or give us read access. We count pages and macros, read a sample of the hardest pages, and send a fixed quote.
  2. Inventory and map. Every page gets a decision: keep, merge, rewrite or retire. We propose the new sidebar and the macro mapping, and you approve both.
  3. Conversion. We run the conversion, then edit by hand: tables that lost their structure, headings used as formatting, and content buried in expand macros that deserves a page of its own.
  4. Redirects and launch. For a public space, retired Confluence pages get short pointer pages to the new site, and the space is archived once traffic has moved. Where the docs sit on a domain you control, permanent redirects run on your host.
  5. 30-day fix window. For 30 days after delivery, we fix anything we got wrong, from a broken macro conversion to a missing attachment, at no charge.

If your team keeps editing Confluence during the project, we track changes made after the export and bring them across before launch, so there is no content freeze and no lost work. Before the full run, we also convert a few of the hardest pages you choose, so you see what the macro mapping does to your real content before approving it.

Pricing#pricing

Up to 150 pages
From $4,500
about €4,150
Up to 500 pages
From $9,500
about €8,750
Above 500 pages
Custom quote

Confluence projects use the same page-count tiers as every migration. The tier is set by the page count we agree during scoping, and pages we decide to retire do not count toward it. Above 500 pages, we quote after reading a representative sample.

What moves the price in Confluence specifically: heavy macro use, especially includes, excerpts and third-party macros; spaces where the page tree and the permissions disagree about what is public; content spread across several spaces that have to merge into one site; and attachments that are really documents in their own right, such as PDFs and Word files that need converting too. Plain pages with headings, lists, tables and code blocks sit at the easy end of a tier.

Prices are in US dollars; the euro figures are approximate conversions for reference. The general documentation migration page covers other sources and target platforms, including MkDocs and GitBook.

Frequently asked questions#faq

Do we have to move everything out of Confluence?

No. Most teams keep internal pages such as runbooks, specs and meeting notes in Confluence and move only customer-facing and developer documentation. The inventory draws that line page by page, and you approve it before anything is converted.

What happens to Confluence macros?

Each macro type gets a mapping that we agree before conversion. Panels become admonitions, expand macros become collapsible sections, and includes become shared MDX partials. Dynamic macros such as Jira issue lists become static content or links, and we flag every one so nothing disappears silently.

Can non-engineers still edit the docs after the move?

Yes, with a little setup. Writers can edit in the GitHub or GitLab web editor, or in a Git-based CMS if you want one. The contributor guide covers the path your team will actually use, and we walk them through it before handoff.

Do page comments and version history come across?

Page content, attachments and structure come across. Inline comments and per-page history stay in Confluence, so keep the space archived if you need that record. From launch on, Git keeps a full history of every change.

Does it matter whether we use Confluence Cloud or Data Center?

Not much. The export paths differ slightly, but the conversion, cleanup and redirect work is the same on both.

Send us what you have

Send a link to your docs, an export, or a few lines on what needs to change. We reply within 1 business day with questions or a quote.